162 ALASKAN GLACIER STUDIES
east margin and a small area of terminal deposits described below, the ice surface is so free from debris that it appears clear in the photographs. Examined in detail one finds a small amount of moraine material scattered over the surface but not in quantity sufficient to form a continuous sheet.
The large proportion of clear ice in the Fourth Glacier, which resembles the condition of the Hidden, Nunatak, Orange, Hubbard, and Turner Glaciers, as contrasted to the d6bris-covered surfaces of such glaciers as the Butler, Haenke, Black, Galiano, Atrevida and Lucia, presents an interesting problem. Why should some of the glaciers have the condition illustrated in the Fourth Glacier and others the condition found in the Atrevida? In seeking to answer this question conclusively one would probably need to know more about the upper ice conditions in the larger glaciers, and whether the glaciers are actively moving, like the Hubbard, or slowly moving, like the uncrevassed Fourth Glacier. It is to be noted,however, that it is the larger glaciers that have the least amount of d6bris-cover. It is possible that the absence of debris is dependent upon the breadth of the valley, but under normal conditions the breadth of such a valley should diminish in the upper portions so that from the tributary sources, at least, such large quantities of de*bris would be supplied that it would form an ablation moraine over the wasting glacier end. The mere size of the valley in. which the glacier ends does not, therefore, seem a sufficient explanation of the absence of ablation moraine in a region where so many glaciers have such extensive morainic cover. Only one hypothesis occurs to us which satisfactorily explains all the conditions, and this is that such glaciers as the Fourth, Hidden, and Orange, although large, receive their main snow and del>ris supply, not from the union of numerous small tributaries decending through narrow mountain valleys, but from snowfall upon broad ice divides and the avalanching of snow and rock on the mountain sides which enclose these through glaciers, with, of course, the addition of some small, short, relatively ineffective tributaries of the cascading type. By this explanation it would seem impossible for dlbris to extend in large quantities far enough out over the ice to furnish the material for a continuous cover of ablation moraine, while in such narrow valley glaciers as the Atrevida, debris would find its way out even to the center of the glacier. The absence of d6bris-cover on such active, clear-ice glaciers as the Hubbard, to which many narrow valley glacier tributaries doubtless contribute ice, may be due to the failure of ablation to lower the broken surface far enough to concentrate the debris by the time the iceberg-discharging sea cliff is reached.
Three facts indicate that the Fourth Glacier is not now in a very active stage. These are, first, its smoothness, second the fact that its front does not extend to the mouth of the mountain valley, and third, the evidence that it is now rapidly receding. Its front lies well within the mouth of the mountain valley, and there is, therefore, no opportunity for its expansion to form a piedmont ice bulb. In this respect Fourth Glacier differs widely from the Atrevida, Lucia, and Marvine and other tributaries of the Malaspina Glacier.
From the front of the Fourth Glacier two small streams emerge, one from either side. These flow along the front and join a large stream which issues near the center. This medial stream course is determined by a depression between the steeply-sloping alluvial fans which were constructed during a recent stage when the marginal drainage pursued a course down each side of the valley. In the depression Beasley Creek is building a small alluvial fan. It then flows along the east side of the mountain valley to the mouth